About 20 percent of women report having had one miscarriage, and 5 percent have had two or more spontaneous losses. Even after extensive investigations, 30 to 40 percent of recurrent fetal losses remain unexplained. Several investigators have suggested that inherited thrombophilic conditions, such as factor V Leiden, could be associated with recurrent fetal loss. Rey and colleagues reviewed the literature to establish the strength of such an association for the more common inherited thrombophilic disorders.

The authors identified relevant studies from electronic databases, searched reference lists, and contacted authors of various studies for additional data. Abstracts, letters, case reports, editorials, meta-analyses, and reviews were excluded from the study. Recurrent fetal loss was defined as two or more losses during the pregnancy period studied by the investigators for the individual studies; nonrecurrent fetal loss was defined as one loss during a study period. The disorders of interest included factor V Leiden, homozygous methylenetetrahydrofolate reductase mutation, activated protein C resistance, prothrombin G20210A mutation, protein S deficiency, protein C deficiency, and antithrombin deficiency. Each study was independently rated for quality by two of the authors of this study.

Of the studies reviewed, 31 were of appropriate quality for the analysis. These studies varied in definitions, factor or factors studied, outcomes, and methodologies. Factor V Leiden was the most studied factor, with 18 studies of recurrent fetal loss. It was significantly associated with early (before 13 weeks of gestation) and late (after 22 weeks) recurrent fetal loss, even when other potential underlying causes were excluded. Factor V Leiden was also associated with nonrecurrent fetal loss, especially after 19 weeks of gestation.

Weaker evidence suggested significant associations between recurrent fetal loss and activated protein C resistance, protein S deficiency, and prothrombin G20210A mutation. The other thrombophilic conditions were not significantly associated with fetal loss.

The authors conclude that only certain thrombophilic conditions are associated with recurrent fetal loss, and that the association differs for losses occurring early and later in gestation. First-trimester losses are associated with factor V Leiden, activated protein C resistance, and prothrombin G20210A mutation. For the most common factor in most U.S. populations, factor V Leiden, the risk of late recurrent loss is even higher than that of early miscarriage. Factor V Leiden, prothrombin G20210A mutation, and protein S deficiency are associated with nonrecurrent loss late in pregnancy. The mechanism of these associations is believed to involve excessive thrombosis of placental vessels, but much remains to be studied.

editor's note: Thrombophilic disorders are previously “small print” conditions that are rapidly becoming highly significant clinical issues in women's health. Are they necessary but not sufficient factors for clotting disorders in contraception and hormone use as well as fetal loss? Do they need other risk factors such as smoking to trigger their pathologic potential? Will physicians soon screen for them routinely before any treatment or surgery that has clotting potential? Taken to the extreme, will airlines ask factor V Leiden passengers to exercise during flights to avoid thrombosis? As our awareness has increased, the surprise is how common these factors are in some populations. About 10 percent of patients in a typical family practice are affected. In European populations, the prevalence of factor V Leiden may be higher. Besides pregnancy loss, heterozygotes for this factor have a sevenfold increased risk of thromboembolism, and homozygotes have an 80-fold increased risk (See Bombeli T, et al. Prevalence of hereditary thrombophilia in patients with thrombosis in different venous systems. Am J Hematol June 2002;70:126–32, and Kenneth AB. Hypercoagulable states. In: Hoffman R, ed. Hematology: basic principles and practice. 3d ed. New York: Churchill Livingstone, 2000). One wonders what the possible evolutionary advantage could be for the mutation to be so common in European populations. Was the risk of bleeding to death from trauma greater than the dangers of thrombosis in our more physically active ancestors?—a.d.w.